5206 perience greater deshielding effects from proximal anisotropic pyridine molecules coordinated to the OH proton. Subtle conformational differences in ring C between compound 2 and compounds 3 and 5 are indicated from the solvent shift results recorded for the Clo methylene protons (see Table I). Thus, in 2, relative to 3 and 5, both the Clo protons resonate to low enough field (in C5DjN) to be readily identifiable and, i n addition, both Hlogaxial and Hlo, equatorial experience smaller (but similar) solvent deshielding effects (A = -0.44 and -0.35 ppm, respectively) than corresponding protons in 3 and 5 . Since the magnitude of pyridine solvent shifts is distance dependent, lfi these results are in corroboration with the conclusions of the previous section in that, in 2, ring B assumes the other possible conformation such that the Cla methylene protons are almost equidistant from the phenolic oxygen. This is in contrast to the conformation of ring B of compounds 3 and 5 where Hloa equatorial is considerably closer to the phenolic OH (and hence the large solvent shift) than Hlnpaxial. These conclusions are in good accord with the internuclear distance, r, between the phenolic oxygen and Clo protons obtained from tbe Westheimer method. Tbus, in 1 ( Y ~ ~ ~ - -=O ~ 2.30 A) and 3 (rEIloa--Ol= 2.34 A and Y ~ = ~ 3.46 ~ A), - Hlo ~ and HIOaare in close proximity to the phenolic oxygen, whereas in 2 (rHloa-ol= 2.65 A and rHlog-O1 = 2.48 A), HIOaand Hlopare located at nearly equal but intermediate distances from the phenolic oxygen.
Experimental Section I-Ag-THC (1) and I-A*-THC( 5 ) were obtained from R. Mechoulam, Laboratory of Natural Products, School of Pharmacy, The Hebrew University, Jerusalem, Israel. Gas-liquid partition chromatography ( 8 1 ~ indicated ~ ) ~ ~ that the purity of these samples was about 95 %. The 3-n-hexyl analog of A6n(1n1)-THC (2) ("synhexyl") was obtained from Abbott Laboratories. Glpc indicated that the sample was at least 90 pure. I-truns-Hexahydrocannabinol (3)4c was prepared by taking a 313-mg sample of A8-THC( 5 ) in 35 ml of ethanol and hydrogenating over 50 mg of PtOz for 4 hr at 45 psi pressure. Evaporation of the solvent after removal of the catalyst by suction filtration gave 300 mg of a light tan resin. Glpc and tlc investigation indicated that the product was a mixture of two isomers. The molecular weight of HHC (3) was confirmed by high-resolution mass spectrometry. Nmr spectra were recorded using a Varian HR-220 spectrometer. Decoupling and NOE studies were carried out on a Varian HA-100 spectrometer in the frequency sweep mode. NOE effects were measured on nitrogen-sparged solutions (sample concentrations were approximately 8 w/v) with TMS as internal lock and reference, The irradiating audiooscillator was a Hewlett-Packard 200 ABR, and power requirements for NOE studies were determined by slowly increasing millivolt output until area increases were optimized. Each peak indicating increases in signal height was integrated a t least ten times with and without optimum power and NOES calculated from the average values.
z
z
Acknowledgments. We are grateful to Max M. Marsh for stimulating this work and for helpful dis~ cussions. (31) An F & M 810 instrument was used with H? flame detector isothermally at 230". The column was a 6 ft X 0.25 in. SS packed with 3 Z silicone gum XE-60 on 100-120 Chromosorb WAW DMCS. TIC employed silica gel plates (Merck) using 1 : 4 Et?O-hexene with 1 % vanillinHISO, spray for development.
The Stereochemistry of Aminophosphines' A. H. Cowley, M. J. S. Dewar,* W. R.
and W. B. Jennings3
Contribution from the Department of Chemistry, The University of Texas at Austin, Austin, Texas Received February 7, 1970
78712.
Abstract: The proton magnetic resonance (pmr) spectra of a series of aminophosphines, R?NPXY, have been examined over the temperature range 40 to - 150". In the majority of these compounds the nitrogen substituents, R , became diastereotopic at low temperatures. Using the technique of matching the observed and computer simulated line shapes, it was possible to calculate the rates of, and activation parameters for, the implied stereochemical changes. The observed steric deceleration with increasing steric bulk of the nitrogen substituents, together with the inability to observe a barrier in 2,2-dimethyl-l-diphenylphosphinoaziridine, provide evidence that the observed barriers in the acyclic aminophosphines relate to torsion around the phosphorus-nitrogen bond rather than to pyramidal nitrogen inversion. The origins of these barriers are discussed from the standpoints of steric effects, lone pair-lone pair repulsions, and pa-da bonding. In contrast to an earlier report it is found that the symmetrical aminophosphines, R2NPX2,are still undergoing rapid P-N bond rotation on the nmr time scale at -80'. The observation of diastereotopic R groups below - 120", together with steric considerations, suggeststhat the gauchc-type conformation is adopted at low temperatures. The pmr spectra of CsH5As(CL)N(CH3)? have also been recorded over a wide range of temperatures, leading to the measurement of the first arsenic-nitrogen rotational barrier.
here is considerable current interest in the use of nuclear magnetic resonance (nmr) to investigate the stereochemistry of trivalent nitrogen attached to
T
( I ) This work was supported by the Air Force Office of Scientific Rcsearch. through Grant No, AF-AFOSR-1050-67. the National Science Foundatron, through Grant GP-9518, and the Robert A . Welch Foundation. (2) To whom all correspondence should be addressed. (3) (a) Department of Chemistry, The Queen's University, Belfast, N. Ireland. (b) Robert A. Welch Postdoctoral Fellow.
Journul of the American Chemical Society
group V and group VI heteroatoms. This area encompasses hydrazines, aminophosphines,j-* hydroxyl(4) (a) B. H. Korsch and N. V. Riggs, Tetrahedron Lett., 5897 (1966); (b) G. J. Bishop, B. J. Price, and I. 0. Sutherland, Chem. Commim., 672 (1967); (c) A. Foucand and R. Roudant, C. R . Acud. Sci., Paris, Ser. c, 266, 726 (1968); (d) M. J . S. Dewar and W. B. Jennings,J. Amer. Chem. SOC.,91, 3656 (1969); (e) J. R. Fletcher and I. 0. Sutherland, Chem. Commun., 707 (1969); (f) J . E. Anderson, D . L. Griffith, and J. D. Roberts, J . Amer. Chenr. Soc., 91, 6371 (1969); (g) M. J. S . Dewar and W. B. Jennings, Tetruhedron L e t t . , 339 (1970).
92.17 J August 26, 1970
5207
a r n i n e ~ , sulfinamides, ~ and sulfenamides. 11,12 The principal points of interest in these studies have centered around the identification of the rate-determining stereochemical processes, assessment of the conformational preferences at low temperatures, and evaluation of the various factors which might influence the magnitudes of the rotational or inversional barriers. In the case of the aminophosphines the phosphorus and nitrogen atoms each possess a lone pair of electrons; hence the possible stereochemical processes comprise rotation around the phosphorus-nitrogen bond, and pyramidal inversions at both nitrogen and phosphorus. However, phosphorus inversion is a relatively high energy process (see later) which is not expected to occur in the temperature range of concern here. The choice between P-N torsion and pyramidal nitrogen inversion as the rate-determining step is not trivial because barriers to nitrogen inversion can vary over a wide range. Thus, in some cases it has been s h o ~ n " , ~ , ~that ~ , ~ 'the presence of a heteroatom directly bonded to nitrogen substantially raises this barrier, while in others, where the possibility of a pir-dir interaction exists, the barrier is reduced.12*15In our preliminary communication6 it was argued that P-N bond torsion was the rate-controlling feature in acyclic aminophosphines. In the present paper these arguments are strengthened by data for more compounds and by failure to detect a nitrogen inversional barrier in 2,2-dimethyl-l-diphenylphosphinoaziridine down to -150". By extending the measurements to temperatures below - loo", it has also become possible to measure the P-N rotational barriers in symmetrically substituted aminophosphines of the type (CH3)*NPX2. The spectral observations on these compounds are important because they provide new evidence concerning the conformational preference of aminophosphines at low temperature. Another concern of the present paper is an attempt to assess the relative importance of steric effects, lone pair-lone pair repulsions, and pir-dir bonding in maintaining the preferred geometry. Finally, proton nmr spectral data are presented for the aminoarsine, C6HjAs(Cl)N(CH3)2,leading to the measurement of the first arsenic-nitrogen rotational barrier. Experimental Section Elemental analyses were performed by Galbraith Laboratories. All melting point samples were sealed in capillaries under an argon atmosphere. All operations involving the aminophosphines were ( 5 ) M. P. Simonnin, J. J. Basselier, and C. Charrier, Bull. SOC.Chim. Fr., 3544 (1967). (6) A. H . Cowley, M. J. S. Dewar, and W. R . Jackson, J . Amer. Chem. SOC.,90, 4185 (1968). (7) D . Imbery and H. Friebolin, Z . Naturjorsh., 23b, 759 (1968). (8) H. Goldwhite and D . G. Rowsell, Chem. Commun., 713 (1969). (9) (a) D . L. Griffith and J. D . Roberts, J . Amer. Chem. SOC.,87, 4089 (1965); (b) B. J. Price and I. 0. Sutherland, Chem. Commun., 1070 (1967); (c) M. Raban and G. W. J. Kenney, Jr., Tetrahedron Lett., 1295 (1969). (10) H. J. Jakobsen and A. Senning, Chem. Commun., 617 (1967). (11) M. Raban, G . W. J. Kenney, Jr., and F. B. Jones, Jr., J . Amer. Chem. SOC.,91, 6677 (1969), and references therein. (12) J. M. Lehn and J. Wagner, Chem. Commun., 1298 (1968). (13) (a) S . J. Brois, J . Amer. Chem. SOC.,90, 506 (1968); (b) S . J. Brois, Tefrahedron Lett., 5997 (1968). (14) F. A . Johnson, C. Haney, and T. E. Stevens, J . Org. Chem., 32, 466 (1967). (15) F. A. L. Anet, R . D . Trepka, and D. J. Cram, J . Amer. Chem. SOC.,89, 357 (1967).
carried out either under an inert atmosphere or by standard vacuum line techniques. Materials. 2,2-Dimethylaziridine, bp 69-70' (760 mm), was prepared by the method of Cairns.16 Methylamine-N-dz was prepared by the exchange of CHINHZwith Dz0.'7 The exchange reaction was allowed to proceed until the deuterium content of the amine was greater than 90% on the basis of nmr and mass spectroscopy. The chlorophosphines CF8PC1, and (CF&PCI were obtained from the reaction of the corresponding iodophosphines with HgClZ.la The other starting materials were procured commercially and used without further purification. Trichlorofluoromethane ("Freon-11") and dichlorofluoromethane ("Genetron21") were obtained from Matheson and had purities of 99.9 and 99.0%, respectively. The following aminophosphines were prepared by slow addition of the appropriate quantity of the amine in ether solution to a stirred ethereal solution of the halophosphine at a temperature below - lOO.19 The stirred solution was allowed to assume ambient temperature and stirring was maintained for a few hours thereafter. The amine hydrochloride was then filtered off, the solvent stripped off under reduced pressure, and the crude product purified by fractional vacuum distillation Chloro(dimethylamino)phenylphosphine, bp 47-50" (0.25 mm) ( k z obp 79" (2.5 mm)). Dimethylamino(diphenyl)phosphine, mp 32-33 '; bp 96.7" (0.015 mm) (lit.21mp 31.5-33.5"; bp 123-124" (0.1 mm)). Dichloro(dimethylamino)phosphine, bp 57-59' (24 mm) (lit.22 bp 150" (760 mm)). Chloro(diethylamino)phenylphosphine, bp 146-148 (2 mm) ( M Z 3bp 82-84" (0.05 mm)). Diethylamino(diphenyI)phosphine, bp 125-127" (0.07 mm) (lit.23 bp 126" (0.1 mm)). Dichloro(dimethylamino)phosphine, bp 75-78 a (22 mm) ( M Z 2 bp 78" (17 mm)). Chloro(dibenzylamino)phenylphosphine, bp 170-175 O (0.05 mm), mp 47-49'. Anal. Calcd for CzoHIKINP: C, 70.7; H, 5.6; CI, 10.4; N, 4.1; P, 9.1. Found: C, 70.4; H, 5.65; C1, 10.0; N, 4.0; P, 9.5. Chloro(diisopropylamino)phenylphosphine, bp 87-88 O (0.05 mm) (lit.7 bp 122" (0.2 mm)). A m l . Calcd for C12HlqC1NP: C, 59.1; H, 7.8; C1, 14.85; N, 5.7; P, 12.7. Found: C, 59.0; H, 7.8; C1, 14.6; N, 5.6; P, 12.6. Chlorobis(dimethylamino)phosphine, bp 68" (11 mm) (lit. 2o bp 64" (10 mm)). 2,2-Dimethyl-l-diphenylphosphinoaziridine,bp 120-124 O (0.2 mm), mp 48-50". Anal. Calcd for C1eH18NP: C, 75.3; H, 7.1; N, 5.5. Found: C, 75.2; H, 7.2; N, 5.3. Chloro(dimethylamino)phenylarsine, bp 49-50" (0.05 mm). Anal. Calcd for CaHI1AsCIN: C, 41.5; H, 4.8; C1, 15.3. Found: C,41.65; H,4.9; C1, 15.2. The volatile aminophosphines (CH&NP(Cl)CF3, 2 4 (CH& NP(CF3)2,Z5 and CH3N(H)P(CF3)2z5were prepared by published procedures and fractionated by trap-to-trap distillation until their vapor pressures conformed to the literature values. The N-deuterio compound, CH3N(D)P(CF& was prepared by a modification of Harris' method25 in which CH3NH2 was replaced by CH3NDz. The vapor pressure was 26.4 mm at 0". Nmr Spectra. All spectra were determined on a Varian Associates HA-100 spectrometer equipped with a variable-temperature accessory. Probe temperatures down to -60" were calibrated against methanol spectra as described in the Varian Users Manual. Temperatures below -60" were measured by inserting a copperconstantan thermocouple into a sample tube containing 0.5 ml of a mixed isohexanes solution. The thermocouple was calibrated before each experiment using the boiling point of water and the sub(16) T. L. Cairns, ibid., 63, 871 (1941). (17) H. Wolff and A. Hopfner, Ber. Bunsenges. Phys. Chem., 69, 710 (1965). (18) F. W. Bennett, H . J. Emelkus, and R. N. Haszeldine, J . Chem. SOC.,1565 (1953); A. B. Burg and J. F. Nixon, J . Amer. Chem. Soc., 86, 356 (1964). (19) This is essentially the procedure described by W. A. Hart and H. H . Sisler, Inorg. Chem., 3, 617 (1964). (20) H. Noth and H . J. Vetter, Chem. Ber., 96, 1109 (1963). (21) L. Maier, Helu. Chim. Acta, 46, 2667 (1963). (22) A . B. Burg and P. J. Slota, J . Amer. Chem. Sac., 80, 1107 (1958). (23) G. Ewart, D. S . Payne, A. L. Porte, and A. P. Lane, J . Chem. SOC..3984 (1962). (24) J. F: Nixon and R. G. Cavell, ibid., 5983 (1964). (25) G. S . Harris, ibid., 512 (1958).
Cowley, Dewar, Jackson, Jennings
1 Aminophosphines
5208 Table I. P-N Bond Rotational Barriers and Spectral Data for Chloro(dialkylamino)phenylphosphines, CeHaP(C1)NRn
R
Temp, "C
Solvent
CH 3
CFCL
30
- 80 CHzCH3
CFC13-CDC13
CHzCeHs
30
- 80
(1:l)
CDC13
30
- 70 CH(CH3)z
CFCla
30
- 50 CH(CH,)CHzCH3Q
csz
Other JPKCH, couplings, Hz Hz
TCHs'
7.40 7.22 7.75 8.92 t 8 . 7 4 t, 9 . 0 5 t 5.91 qd 5.72 q, 6 . 2 8 qd 8 . 7 1 d, 8 . 9 1 d 8 . 5 5 d, 8 . 6 3 d, 8 . 8 9 d, 9 . 1 7 d 8.7
12.6 19.2 6.7 12.0
e
e 10.0
Tcoaiea, "C
Exchange A F t , c rate,b kcall sec-' mol-'
- 50
104
10 9
JHCH = 7.1
- 57
50
10 8
JHCH = 14.4 JHCH = 15.8 JPNCH = -0.5 JHCCH = 6.7
-46
f
-15 (-lo)*
70 105 250
(-5Y 15
12.8 12 9 12.7 14 60
d = doublet; t = triplet; q = AB quartet. Exchange rates were calculated at the coalescence temperature using a computer simulated line shape analysis procedure; the program was of the many site type (see text). Derived from the Eyring equation assuming a transmission Spectra were observed with irradiation at the 31Pfrequency. f Not determined coefficient of unity. Methylene proton absorptions. owing to the complexity of the spectra. 0 Values taken from ref 7. Additional temperature at which the exchange rate was determined. ti
Table 11. P-N Bond Rotational Barriers and Spectral Data for Aminophosphines, RzNPXY
R
Y
X
Solvent
Temp, "C
JPNCH, TCH~
HZ
Other couplings, Hz
Tcoaies, "C
Exchange rate, Afl,a sec-' kcal/mol ~~
c1
c1
CHFClz
CF3
CF3
CFiClz
-74 -125
-80
7.12 7.06 7.10 7.07
12.4 19.2 4.9 8.9
JFCXPCH
-113 (- 107)*
14" 35 c
8 4 8.3
- 10Sd
15d
8 7
8.5d 4OC 34
8.8 8.9
0.5 -142
c1
CF,,
CFC13
7.03 7.10 7.07
-14 -4 11.2
-80 -80
6.98 7.11 7.48 7.25 9.08 8.81
18.1 5.0 9.8 9 9.5 13.0
-120
7.26
12.3
30 -100
CsHj F CsHj
c1
CsH5 F CsHj
c1
CHFClz CHFClz CHFClz CFC13-CDCl3
-130 -108
(-107)*md (- 96)*
JFCPNCH - 70 0.5 - 75d
JFPNCH 4 JHCCH 7.0 JHCCH 7,2
-30
10.5 -10.2
